MODELLING OF DOWNDRAFT GASIFIER USING COMPUTATIONAL FLUID DYNAMICS SOFTWARE-

FLUENT

A thesis submitted for the degree of Master of Science

Energy Systems & Environment

Noorhelinahani Abu Bakar June 2003

Mechanical Engineering Department University of Strathclyde in Glasgow

ACKNOWLEDGEMENT

I would like to express my gratitude to Mr Craig Mclean for his guidance and advice

during this dissertation and to Professor J.A Clarke for his expert knowledge and

guidance throughout the course of the project.

I also wish to express my thanks to all the lecturer and staff of the Energy Systems

Research Unit (ESRU) for the helping throughout this one-year MSc course.

Finally, my sincere thanks to my friends Nadiahnor Mohd Yusop & Abd Rahim from

Leeds University and Kamarul Ariffin from Cranfield University for their continual

help in learning FLUENT software and support during my dissertation and until now.

i

ABSTRACT

Biomass energy combined with computational fluid dynamics (CFD) software

have been described in some detail in the following study. The aspect of biomass

gasifier has been investigated with regard to the potential of designing future gasifier

using simulation in FLUENT software.

The project is about the modelling of fluid flow of biomass fuel inside biomass

downdraft gasifier using FLUENT software. Using the typical design of gasifier, the

model is produced in 2-Dimensional for easier iterations and simulations.

Two types of gasifier are measured; pressurised and atmospheric gasifier. The

gasifier’s model is set to have initial temperature at 800° C for both types of gasifier

and 5.0 MPa for pressurised gasifer.

After iterations in FLUENT software with first-order discretization, the results

obtained are convergent and show that using second-order discretization is not

suitable for 2-Dimensional model.

Finally, future investigations are also suggested in this study for more accurate

result using computational fluid dynamic (CFD) software.

ii

TABLE OF CONTENTS

PAGE

ACKNOWLEDGEMENT ..i

ABSTRACT ..ii

CHAPTER 1 - INTRODUCTION ..1

1.1 BIOMASS RESOURCES ..1

1.2 SIGNIFICANCE OF CURRENT WORK ..2

1.3 BIOMASS GASIFIER ..3

1.4 OVERVIEW OF PROJECT TOPIC ..5

1.5 CONCISE PROJECT DEFINITION: MOTIVATION FOR ..6

CURRENT WORK

1.6 AIM AND OBJECTIVES OF THE STUDY ..7

CHAPTER 2 – THEORY AND LITERATURE REVIEW ..8

2.1 INTRODUCTION ..8

2.2 THEORY OF BIOMASS GASIFICATION ..8

2.2.1 CONCEPT OF GASIFICATION AND PYROLYSIS ..8

2.2.2 GASIFICATION EQUATION ..10

2.3 MODEL ..11

2.3.1 CHARACTERISTICS OF BIOMASS GASIFIER ..11

2.3.2 GAS PRODUCTION FROM GASIFIER ..12

2.4 COMBUSTION REACTION MECHANISMS ..14

2.5 THEORETICAL CONSIDERATIONS ON CFD ..16

2.5.1 PRE-PROCESSOR ..16

2.5.2 SOLVER ..17

2.5.3 POST-PROCESSOR

2.6 PROBLEM SOLVING ..18

CHAPTER 3 – METHODOLOGY ..19

3.1 INTRODUCTION ..19

3.2 GASIFIER MODELLING USING CFD SOFTWARE ..19

3.2.1 FLUENT 5.5 ..19

3.2.1.1 PRE-PROCESSING ..19

3.2.1.2 SOLVER ..20

3.2.1.3 POST-PROCESSING ..20

3.3 MODEL SETTING ..21

3.3.1 DIMENSION ..21

3.3.1.1 DOWNDRAFT GASIFIER ..23

3.3.1.2 PRESSURISED DOWNDRAFT GASIFIER ..26

3.4 MODELLING PROCEDURE ..27

CHAPTER 4 - RESULT ANALYSIS AND DISCUSSION ..28

4.1 INTRODUCTION ..28

4.2 MODEL DESCRIPTION ..28

4.3 RESULT ANALYSIS FROM DOWNDRAFT GASIFIER ..30

4.3.1 FIRST-ORDER DISCRETIZATION ..30

4.3.2 SECOND-ORDER DISCRETIZATION ..31

4.4 DOWNDRAFT GASIFIER CONTOUR RESULTS ..33

4.4.1 PROFILE FOR FIRST-ORDER DISCRETIZATION ..33

4.4.1.1 TEMPERATURE & PRESSURE CONTOUR ..34

4.4.1.2 TEMPERATURE & PRESSURE DISTRIBUTION ..36

4.5 RESULT ANALYSIS FROM PRESSURISED ..39

DOWNDRAFT GASIFIER

4.5.1 FIRST-ORDER DISCRETIZATION ..39

4.5.2 SECOND-ORDER DISCRETIZATION ..39

4.6 PRESSURISED DOWNDRAFT GASIFIER CONTOUR RESULTS ..40

4.6.1 PROFILE FOR FIRST-ORDER DISCRETIZATION ..40

4.6.1.1 TEMPERATURE & PRESSURE CONTOUR ..40

4.6.1.2 TEMPERATURE & PRESSURE DISTRIBUTION ..42

CHAPTER 5 – CONCLUSION & RECOMENDATIONS ..46

REFERENCES ..48

APPENDICES ..50

APPENDIX 1.1: LIST OF GASIFIER ..50

APPENDIX 1.2: SUMMARY - FIRST-ORDER DISCRETIZATION ..52

FOR DOWNDRAFT GASIFIER.

APPENDIX 1.3: SUMMARY - FIRST-ORDER DISCRETIZATION ..60

FOR PRESSURISED DOWNDRAFT GASIFIER.

Chapter 1 – Introduction

1.1 Biomass Resources

The use of biomass fuels is rapidly increasing in northern, mid and Eastern

Europe. According to European Commission in the white papers’ community strategy

& action plan in the policies of the European Union (EU), biomass is expected to play

a major role as a renewable energy source1 (Ronald, 2001).

According to F. Dumbleton, in the UK there are about 14 000 000 tonnes/year

of straw that can be obtained from the production of wheat, barley, oats, oilseeds rape

and linseed. From this amount, it is estimated that 4 000 000 tonnes/year could be use

for producing energy. Apart from that 1 000 000 tonnes/year of forest residue are

produced and it is raised up to 2 000 000 tonnes/year in the next 20 years. For woods

it is estimated about 960 000 tonnes/year as logs2.

Many advantages can be obtained from biomass energy such as clean burning

fuel that contributes significantly to the world energy supply. The potential for

greenhouse gases production is also reduced using the biomass energy. Using gasifier

as a medium for the production of syngas from biomass makes lots of benefit for

people, as the sources are abundance.

It can be made available in a variety of shapes like logs, chips, bales, charcoal,

sludge, fluids, straw, powders and etc. Woodchips is one of the sources of biomass

renewable energy that can produce gas from gasification. It has the greatest potential

of any renewable energy option for base-load electric power production for electricity

generator and also heating.

Recently, there are considerable interest has been shown in the gasification of

wood and its use. However, the gas produce from the wood gasification is low in

calorific value3. Therefore, in order to increase the efficiency several studies can be

carried out. For example perhaps enlarge the gas turbine combustor, or enlarge the

gasifier. As for this dissertation, the effect of pressurised downdraft gasifier is being

1 Ronald, V.S. (2002) How European Waste Will Contribute to Renewable Energy. Vol. 30 pp. 471-475. 2 Dumbleton, F. (2001) Standardisation of Solid Biofuels in the UK , AEA Technology Environment, Report 1, ESTA 32192001. 3 McIlveen, W. et al. (2001), Bioresource Technology, A re-appraisal of wood-fired combustion, Vol. 76, pp 183-190

1

studied and the results are compared with the atmospheric gasifier for optimum

efficiency.

1.2 Significance of current work

There are several types of gasifier that are normally used for converting

biomass fuel to get power. For example, updraft gasifier, downdraft gasifier and

fluidised bed gasifier. (Refer Appendix 1.1 for an explanation of each gasifier)

In the dissertation, wood chips gasifier is modelled in downdraft gasifier

which is based on experimental set-up data from pilot scale downdraft gasifier study

that have been done by Dogru et al (2002). Instead of wood chips, the hazelnut shell4

was used for the fuel in the gasifier. This study has shown that the quality of the

product gas is found to be dependent on the smooth flow of the fuel and the

uniformity of the pyrolysis process.

They also suggested that gasification of shell waste products is a clean alternative

to fossil fuels and this product gas can be directly used in internal gas combustion

engines. Therefore, warranting further investment by authorities to exploit this

valuable resource of renewable energy. This pilot gasification system that was

constructed in Mechanical Engineering Lab in New Castle University provide

reasonable base like input data for pressure or temperature and also the dimension of

the downdraft gasifier which can be a core for the modelling using the computer

software.

As expected, the model can provide useful predictions of operating other

downdraft gasifier using diverse biomass feeds as fuel. There is consideration that has

to put in mind about the environmental impact, from indirect and direct emissions

release from the combustion process. For this dissertation, universal emission-

releasing gas model based on biomass fuel is not evaluated although it is very

necessary to analyse the product gas from the combustion process. This is because

each case-dependent emissions associated with biomass gasification practice is quite

tedious and time-consuming. Thus, it is necessary in future to model the pollutants

4 Dogru, M. and et al (2002), The International Journal, Gasification of hazelnut shells in a downdraft gasifier Vol. 27 pp 415-427.

2

gas so that it can be used as reference when any gasifier are to be built on local

community or society. So that they can be modelled based on simulation and can be

evaluated in advance, as the result can be demonstrated from modelling using

computational fluid dynamic software.

1.3 Biomass Gasifier

There are many types of gasifier’s reactor that can be used to generate heat from

the chemical process to get the gas for power generation. For example, the fixed bed

reactor. This reactor is used with downdraft gasifier. Downdraft gasifier has four

reaction zones which are drying, pyrolysis, oxidation, and reduction zone from top to

bottom of the gasifier, respectively. This type of gasifier is operated when air is being

drawn from biomass feedstock or hopper at the bottom of the gasifier to the drying

zone.

In the drying zone, biomass fuel descends into the gasifier and moisture is

removed using heat generated in the zones below by evaporating. After that, the next

zone is pyrolysis. In this zone irreversible thermal degradation of dried biomass fuel

descended from the drying zone takes place using the thermal energy released by the

partial oxidation of the pyrolysis products.

In oxidation zone, the volatile products of pyrolysis are partially oxidised in the

exothermic reactions resulting in a rapid rise of temperature up to 1200°C in the

throat region. The heat generated is used to drive the gasification reactions. In the

reduction zone often refer as gasification zone, the char is converted to the producer

gas by reaction with the hot gases from upper zones. The gases are reduced to form a

greater proportion of H2, CO, CH4. While leaving the gasifier at the temperature

between 200 and 300°C, the produced gases carry over some dust, pyrolytics products

(tar), and water vapour.

3

Picture 1.1 Fixed bed gasification reactor.

1. Wood feeding

2. Supply zone-Drying zone

3. Pyrolysis zone

4. Gasification zone

5. Cracking zone

6. Fluid bed after gasification

7. Gasifier

8. Cyclone burner

9. Gas engine

10. Furnace

11. Dust separator

12. Wet scrubber

13. Settler

14. Water filter

15. Gas tank

The picture above shows an example of gasification system that using fixed bed

reactor type that is finally connected to gas tank for next technology such as

production of electricity or heat.

In a downdraft gasifier, the fuel and the producer gas flow downwards through

the reactor enabling the pyrolysis gases to pass through a throated hot bed of char

which is around 1100°C. Thus it will crack most of the tars into light chain

hydrocarbon and water vapours. In addition, the air or oxygen is usually admitted to

the fuel bed through intake nozzles from throat causing pyrolysis to charcoal and

4

volatile that partially burn as they are produced. The gaseous products of this ‘flaming

pyrolytic combustion’ then consume the charcoal, produced during the pyrolysis and

are converted into combustible gases5 (Dogru et al, 2002).

1.4 Overview of project topic

Nowadays, lots of computational fluid dynamics (CFD) simulations were

conducted to study the effects of moisture in biomass, particle injection location, the

burnout of different biomass particle sizes and many others characteristics of the

biomass so that the prediction of the biomass combustion can be made.

This dissertation is useful for future in designing gasifier. A study of fluid

flow inside the gasifier using computational software is done because experiences

with computer simulation are quite insufficient and furthermore the know-how in this

field is almost blank. Most of previous studies are done for coal as for fuel. Definitely

coal gasifier and biomass gasifier are different things as coal char in gasification

reactivity is significantly different from biomass reactivity.

Most manufacturers of gasifier of the actual size are small or medium sized

enterprises with very small resources for the development of new products, up till

now mainly been based on trial and error methods. But existing software is now so

powerful that they can be useful tools in the development of new and environmentally

friendly combustors.

This study which is based on experimental study, the gasifier is modelled using

computational fluid dynamic software, FLUENT 5.5 and the results are compared

with computational simulations.

5 Dogru, M et al. (2002), Fuel Processing Technology, Gasification of sewage sludge using throated downdraft gasifier and uncertainty analysis, Vol. 75, pp. 55-82.

5

1.5 Concise project definition: motivation for current work

Product gas from biomass gasification can be used in many applications such as

in industrial or residential area. Most of this energy production application are

decentralised and coupled with gas turbine or engine and also boiler. A solid fuel fired

for gas turbine combustor should have a high intensity and produce complete carbon

utilisation. Besides that it should maintain fine fly ash particle size and also have low

pressure drop and good turndown.

The technologies for primary conversion of biomass for electricity combustion

involve these processes which are direct combustion, gasification and pyrolysis. In

direct combustion, the oxidation of biomass takes place with excess air. It will

produce hot flue gases which are used to produce steam in the heat exchange sections

of boilers. The steams produce from the combustion is used to produce electricity in a

Rankine Cycle. Normally, only electricity is produced in a condensing steam cycle

while electricity and steam are co generated in an extracting steam cycle.

As for gasification cycles, biomass is partially oxidised by stoichiometric

amounts of oxygen normally with the present of steam. Then it will provide energy

for thermal conversion of the remaining biomass to gases and organic vapours. For

power production, the cleaned gasification product gases will be led directly to a

boiler or the combustion section of an industrial or turbine.

As for pyrolysis, it uses indirect heat to convert solid organic materials, into

gases and liquids. Conversion is carried out in a reaction chamber where air (oxygen)

is excluded. The material within the reaction chamber is heated to temperatures of

between 400 - 800°C usually by combusting a portion of the gaseous or liquid

products. The resultant products of the pyrolysis process always include gas, liquid

and solid char. Operating conditions and reactor design can be selected to control the

relative proportions produced from the process.

According to Richard et al (1996) in indirect gasification cycles external heat

source is used instead of oxygen, that provides the energy for high-temperature steam

gasification of the organic fraction of biomass to vapours and gases6. Therefore,

temperature is very important as the sources of the combustion engine from the

product gas. 6 Richard, L.B. (1996), Biomass-Fired Power Generation in National Renewable Energy Laboratory Conference, New York: Engineering Foundation.

6

1.6 Aim and objectives of the study

The research objective of this study is to develop a 2-Dimensional Gasifier

Model for Biomass Gasification using FLUENT 5.5 Computational Fluid Dynamics.

The overall aim is to study the validity of two types of wood gasifier and the different

profile between pressurised gasifier and non-pressurised gasifier.

The numerical simulation model presented in this study is aimed at direct

combustion in biomass gasifier. This gasifier offers a number of characteristics when

such as intense burning, enhanced heat transfer, capability of improving pressure and

in addition, some potential for improvements to effectiveness and efficiency

especially in gas turbine for power production.

Using the CFD commercial software, FLUENT 5.5, the study will continue to

seek to understand the combustion mechanism of biomass fuel particularly woodchips

in gasifier and then use this at larger scale to advantage for devices where fast

accurate hot gas stream are needed in industrial and home appliances. In particular, to

seek the fluid dynamics of the flow which is heavily dependent on the outlet channel

characteristics.

In the simulation part of the project, a model of downdraft gasifier consistent

with main observed phenomena is proposed. To adapt the model for numerical

analysis, two methods are presented. In the first method, as a preliminary approach,

the combustion process of laminar and steady state of partly hot-wall in the gasifier is

simulated. In the second method, the same geometry but different boundary

conditions for pressurised gasifier, of laminar and steady state combustion process is

simulated and analysed.

Some results obtained from both numerical methods are compared with the

experimental work carried out by M. Dogru et al (2002) and D.J. Purnomo (1990). On

this basis, the validity of proposed model is being discussed.

7

Chapter 2 – Theory and Literature Review 2.1 Introduction In this chapter, basic theory on biomass gasifier is explained and approaches that have

been used previously on gasifier are also discussed.

2.2 Theory of biomass gasification

2.2.1 Concept of gasification & pyrolysis

The history of gasification begins long time ago during Second World War

where people used gasifiers to operate cars, trucks, boats, trains and electricity

generators in Europe. It was approximately about one million of gasifier were

employed but recent technology performed better designs of this gasifier.

Pioneering work on coal fired downdraft furnaces was done in the late 1940’s.

After that, downdraft combustion of wood-chips has been investigated by Aerts &

Ragland (1989) and then biomass gasifier has been reviewed by Reed & Das (1988)7.

According to Xiu et al (2002) since 1960’s there has been extensive development and

deployment of fixed bed gasifier generator set which have grown in size from 60 to

200 kWe output8.

Basically for biomass gasifier, the processes involved are gasification and

pyrolysis processes. Gasification is a partial oxidation process in which organic

materials in biomass are converted to a mixture of gas phase, energetic compounds,

such as carbon monoxide, hydrogen and methane. Carbon is rejected to form char

which is oxidised within the process to produce additional gases, an ash residue and

the heat energy necessary to drive the reactions. It is archived by thermally degrading

organic materials in the absence of adequate air to support full combustion9.

(Heermann et al, 2000).

Apart from that, pyrolysis is another process that occurs in the absence of

oxygen. It is a thermal degradation of carbonaceous materials at temperature between

400 and 800°C either in the complete absence of oxygen, or with such a limited

7 Purnomo, D.J.A & Ragland, K.W. (1990), Twenty-Third Symposium (International) on Combustion, Pressurised combustion of woodchips, pp 1025-1032. 8 Xiu, L.Y. et al (2002), Biomass & Bioenergy, Design and operation of CFB gasification and power generation system for rice husk, Vol. 23, pp 181-187. 9Heermann, C et al. (2000) Pyrolysis & Gasification of Waste. 2nd ed. Juniper Consultancy Services Ltd.

8

supply that gasification does not occur to any appreciable extend. Such processes de-

volatilise and decompose solid organic materials by heat; consequently, no

combustion is possible.

Pyrolysis mechanisms typically include a superposition of kinetic mechanism

for the individual component of the biomass material. The validity and the

development of these mechanisms were being investigated10 by Miller and Bellan

(1997). This is because a reliable description of the global kinetics biomass pyrolysis

is very significant in the developing of biomass thermo chemical conversion systems

that used high temperature chemistry.

These accurate chemical models of biomass gasification and combustion can be

used as input in systems analysis models to adjust engineering parameters such as gas

velocities, particles residence times, and heat transfer rates to optimise the design and

performance of developing technologies.

Purnomo and Ragland (1990) have done a study on pressurised downdraft

combustion of woodchips11 by experimental approaches. They used 2 cm of yellow

poplar chips. They found that pressure drop was about 10% in the gasifier bed and

they found that good agreement between numerical and experimental result at 1-5 atm

of the pressure. Therefore these ranges of data were inserted during simulation setting

in order to validate the result.

10 Miller, R.S & Bellan, J. (1997), Combustion Science Technology, A generalised biomass pyrolysis model based on superimposed cellulose, hemicellulose and lignin kinetis, No. 126, pp 97-137. 11 Purnomo, D.J.A & Ragland, K.W. (1990), Twenty-Third Symposium (International) on Combustion, Pressurised combustion of woodchips, pp 1025-1032.

9

2.2.2 Gasification Equation Biomass is composed of carbon, hydrogen and oxygen, and also nitrogen and

sulphur (refer Table 2.2). Carbon dioxide is obtained from complete combustion of

carbon, and water is obtained from hydrogen. As for oxygen, it is incorporate into

combustion products. In the combustion zone of the gasifier, heat is generated to drive

the gasification reactions as can be illustrated from the equation below.

Wood is burnt from direct combustion with insignificantly amount of oxygen

inside the gasifier. By the conservation of mass, the mass flow remains constant

which means total mass of products equals total mass of reactants. This reactant are

chemically different the product and the products leave at a higher temperature12.

C + CO2 = 2CO2 ------------------------------- (1)

C + H2O = CO + H2 ------------------------------- (2)

C + H2 = CO + H2O ------------------------------- (3)

C + 2H2 = CH4 ------------------------------- (4)

C + 3H2 = CH4 + H2O ------------------------------- (5)

Equation 1 and 2 are the main reactions on the reduction stage and require heat.

Equation 3 is called water-gas equilibrium reaction. Changes in temperature will drive

the equilibrium in either direction. It is by controlling this reaction that the product

gases may be maintained at a constant. Equation 4 and 5 shows the production of

methane gas.

12 Eastop, T.D and McConkey, A. (1998) Applied Thermodynamics for Engineering Technologist, 5th ed, Addison Wesly Longman Limited.

10

2.3 Model

2.3.1 Characteristics of biomass gasifier Gasification for power production involves the de-volatilisation of biomass in an

atmosphere of steam or air or both to produce medium or low calorific value gas. The

syngas which is useful for electricity generation from either turbine, or engine

produced from pyrolysis, is usually a medium heating value (MHV) fuel gas around

14-18 MJ/Nm3 depending on the feed and processing conditions.

Table 2.1 shows different amount of energy produces from biomass fuel which

gives high value of calorific for fuel with less moisture content. The MHV gas

contains a high level of hydrocarbons including uncondensed pyrolysis liquids, and

saturated and unsaturated hydrocarbons, particularly methane, produced by complex

thermal degradation process.

By these calorific values in the table above, all types of biomass fuel are suitable

for feeding to gasifier. Thus, woodchips may also be used in gasification systems to

produce low calorific value combustible gas for clean energy production.

Table 2.1: Energy produces from different types of biomass fuel13

But for this modelling, the effect of the absolute and bulk density, the size and shape

of the particles, moisture content and chemical composition which is proximate and

ultimate analysis and higher heating value play an important role in determining the

fluid flow profile inside the gasifier. Thus, the wood residue chips are being selected

as the availability of density data.

13 LIOR International LV. (2002), Biomass Combustion Technology (7) Wood Lower Calorific Value http://www.lior-int.com/FORMULA/rp7.as

11

Biomass Fuel Net calorific value in dry matter MJ/kg

Net calorific value as received MJ/kg

Sofwood bark 18.5-20 6-9 Birch bark 21-23 7-11 Wood residue chips 18.5-20 6-15 Saw residue chips 18.5-20 6-10 Sawdust 19-19.2 6-10 Cutter chips 19-19.2 13-16 Grinding dust 19-19.2 15-17 Pellets 19.2 16.8

2.3.2 Gas production from gasifier

The production of useful gas from the gasifier are then becoming awareness, as

the green house gasses are releases to the environment. Coal is used as fuel in boilers

as it is very plentiful and cheap but the resources are non-renewable energy. Studies

showed that there are many issues related to the coal and biomass fuel co-firing. But

biomass is more promising energy to the environment as it gives less impact to the

environment.

Figure 2.2 shows the typical ultimate analyses of wood chips and straw compare

with different types of coal. It shows that coal has more production of carbon dioxide

because high content of carbon. As for biomass fuels it has significantly less content

of sulphur thus no amount of sulphur gas releases (Robinson, 1998).

Figure 2.2: The ultimate analysis of biomass and coal fuels14

Wood offer advantages over fossil fuel with regards to gas emission where the

sulphur content of wood is minimal, therefore sulphur dioxide emission are

negligible. As for nitrogen, if temperature of the combustion is being controlled to

avoid the oxidation of the nitrogen from the air, it will not affect the nitrogen dioxide

emission because nitrogen in wood is not temperature dependence. So, the overall 14 Robinson, A et al. (1998) Fireside Issues Associated with Coal-Biomass Co-firing, in Conference on BioEnergy ’98, United State: National Renewable Energy Laboratory

12

nitrogen dioxide will also be low15 (McIlveen et al, 2001). Here are examples of

ultimate analyses of different types of biomass fuel. They have very low value of

sulphur and also low carbon that are more favourable from coal.

Fuel Carbon Hydrogen Nitrogen Sulphur Oxygen Straw 0.45-0.494 0.06 0.004-0.066 0.0012 0.43-0.442Wood without bark 0.47 0.063 0.0016 0.0002 0.46 Bark 0.47 0.054 0.004 0.0006 0.4 Wood with bark 0.47 0.06 0.003 0.0002 0.44 Oak wood 0.494 0.063 00 00 0.445 Pines wood 0.499 0.063 00 00 0.438 Triticale 0.434 0.063 0.0131 0.0022 0.4779 Miscanthus 0.46 0.06 0.007 0.002 0.44

Table 2.3: Ultimate analysis of biomass fuels (dry)

It is shown that in the table above, the composition of wood without bark has low

carbon, nitrogen and sulphur which indicate low emission of these gases when

combustion occur.

15 McIlveen, W. et al. (2001), Bioresource Technology, A re-appraisal of wood-fired combustion, Vol. 76, pp 183-190

13

2.4 Combustion reaction mechanisms Nevertheless, the present computational fluid dynamic models are unable to

solve such complex reaction mechanisms due to its complexity. Employing detailed

reaction mechanisms in the CFD calculations will generate a numerically inflexible

problem that it will be difficult to reach converged solution with the present of the

solver in the CFD commercial software.

Brink et al (2001)16 created a model for biomass combustion using single-step

mechanics and turbulence-chemistry interaction with eddy dissipation combustion

model as the air distribution were injected around the bottom of the model. Therefore,

the use of much simple or single-step reaction mechanisms is preferred rather than the

global reaction mechanism.

In this model several number of simplifying assumptions has been employed for

ease of the discussion. Theoretically, the combustible gas is expected to distribute

uniformly throughout the space into the downdraft gasifier.

Several number of simplifying assumptions has been employed for ease of the

discussion. The main assumptions of the model are as follow:

• Low Mach number for the combustion approximation

• Single-phase gaseous flow of ideal gas

• Neglect the electromagnetic radiation

• Simplified chemistry i.e. single step, Arrhenius

• Newtonian fluid flow; Fourier’s and Fick’s law for the diffusion heat and mass

transfer

The assumption of the low Mach number implies that the ambient flow speeds are low

that the heat release rate is not so strong i.e. no blast waves. The density variations are

due to the differences of the density of the reaction species and the chemically

reaction heat release rate. With the assumptions made, the mathematical model

becomes as illustrated below:

16 Brink, A et al. (2001), IFRF Combustion Journal, Modelling Nitrogen Chemistry in the Freeboard of Biomass-FBC, pp 1-14.

14

• Conservation of mass

(2.3) 0=∂

∂+∂∂

ii

vxt

ρρ

• Conservation of momentum

(2.4) x

Re1

x ijj

1

i

τρρ∂∂+

∂∂−=

∂∂+

∂∂

ijii x

pvvvt

• Conservation of energy

(2.5) 11x 1k

∂∂

−+

∂∂

∂∂=

∂∂+

∂∂ ∑

=

N

Kkkk x

YhScx

hx

hvht α

αα

α σµ

σµρρ

• Conservation of state

(2.6) spacein uniform WYTR R N

10

00 === ∑

=α α

αρρW

Tp

• Conservation of species

(2.7) DY x

Y kk

∂∂

∂∂+=

∂∂+

∂∂

kk xxv

t αααα ρωρρ &

15

2.5 Theoretical considerations on CFD The CFD codes are arranged by the numerical algorithm accordingly, so that the fluid

flow problem can be tackled. There are three main elements of CFD codes in the CFD

packages which consist of pre-processor, solver and post-processor17.

2.5.1 Pre-processor The pre-processor contains all the fluid flow inputs for a flow problem. It can be

seen as a user-friendly interface and a conversion of all the input into the solver in

CFD program. In this stage, quite a lot of activities are carried out before the problem

is being solved. These stages are listed as below:

1. Definition of the geometry

• The region of interests which is the computational domain.

2. Grid generation

• The subdivision of the domain into a number of smaller and non-

overlapping domains. The grid mesh of cells is carried out for the

geometry.

3. Selection of the physical and chemical properties

• The geometry to be modelled.

4. Definition of the fluid properties

5. Specifications of correct boundary conditions

• This is done at model’s cells.

The solution of the flow problem such as temperature, velocity, pressure etc. is

defined at the nodes insides each cell. The accuracy of the CFD solution governed by

the number of cells in the grid and is dependent on the fineness of the grid.

17 Anderson, J. D (1995), Computational Fluid Dynamics, The Basic With Application, International Editions, McGraw-Hill

16

2.5.2 Solver In the numerical solution technique, there are three different streams that form

the basis of the solver. There are finite differences, finite element and finite volume

methods. The differences between them are the way in which the flow variables are

approximated and the discretization processes are done.

1. Finite difference element, FDM

• Describes the unknown flow variables of the flow problem by means of

point samples at node points of a grid coordinate. By FDM, the Taylor’s

expansion is usually used to generate finite differences approximation.

2. Finite element method, FEM

• Use the simple piecewise functions valid on elements to describe the local

variations of unknown flow variables. Governing equation is precisely

satisfied by the exact solution of flow variables. In FEM, residuals are

used to measure the errors.

3. Finite volume method, FVM

• It was originally developed as a special finite difference formulation. The

main computational commercial CFD codes packages using the FVM

approaches involves PHOENICS, FLUENT, FLOW 3D and STAR-CD.

Basically, the numerical algorithm in these CFD commercial packages

involved the formal integration of the governing equation over all the

finite control volume, the discretization process involves the substitution

of variety FDM types to approximate the integration equation of the flow

problem and the solution is obtained by iterative method.

Discretization in the solver involves the approaches to solve the numerical integration

of the flow problem. Usually, two different approaches have been used and once at a

time.

1. Explicit approach

• Usually, this is the most approach that makes sense. It is relatively simple

to set up and program. The limitation is that for a given xt ∆∆ and , must be

17

less than some limit imposed by stability constraints. In some cases,

t∆ must be very small to maintain the stability and consequently long running

time required for the calculation over a given time interval, t.

2. Implicit approach

• For this approach, the stability can be maintained over a large value of

t∆ and fewer time steps required making calculation. Thus resulting less

computer time. Adversely, it is complicated to set up and program. The

computer time per time step is much larger than the explicit approach due

to the matrix manipulation which is required for each time step. This

approach is very accurate to follow the exact transients i.e. the time

variations of the independent variables.

2.5.3 Post-processor

A FLUENT package provides the data visualisation tools to visualise the flow

problem. This includes – vectors plots, domain geometry and grid display, line and

shaded counter plots, particle tracking etc. Recent facilities aided with animation for

dynamic result display and also have data export facilities for further manipulation

external to the code.

2.6 Problem solving In the computational fluid dynamic, using the FLUENT codes provide to solve the

problem numerically. The fundamental involves determining the convergence,

whether the solution is consistent and stable for all range of flow variables.

• Convergence – is a property of a numerical method to produce a solution that

approaches the exact solution of which the grid spacing, control volume size is

reduced to a specific value or to zero value

• Consistent – to produce the system of algebraic equations which can be

equivalent to the original governing equation

• Stability – associates with the damping of errors as a numerical method

proceeds. If a technique chosen is not stable, even the round-off error in the

initial data can leads to wild oscillations or divergence. 18

Chapter 3 – Methodology 3.1 Introduction In this chapter, the main methodology focuses on how to solve the problem using

different type of biomass gasifier which are pressurised and atmospheric condition are

discussed.

3.2 Gasifier modelling using CFD software

3.2.1 FLUENT 5.5

FLUENT 5.5 have been chosen to carry out the simulation for the numerical

simulations. This software has the capability on predicting the fluid flow, heat and

mass transfer, chemical reaction, and related phenomena. This software help

engineers with detailed product development, design optimisation, trouble shooting,

scale-up and retrofitting that can help in reduces engineering cost, while improving

the final design of products in many applications, and one of them is combustion

system design18.

3.2.1.1 Pre-processing

In the first part of numerical simulations, the 2-dimensional geometry of the

pilot biomass gasifier19 and the flow region were created using GAMBIT. 2-

dimensional model has been created because it is easier to model the simulation

comparable to 3-dimensional in terms of computer processor’s time consuming when

running the simulation. Thus, in the setting model, the downdraft gasifier was

examined using 2-D geometry as the asymmetric geometry tends to be diverged when

it was iterated.

Meshing is very important in order to get the solution with high quality results.

Before meshing was done to the model, the model was drawn using unit scale in

GAMBIT and then it was specified or can be modified in metric scale when the

meshing file was exported to the FLUENT 5.5. There are different types of meshing

18 Whalen, P.(2000) Fluent Inc. Releases Version 5.5 of Its FLUENT Software, http://www.fluent.com/about/news/pr/pr17.htm 19 Dogru, M. and et al (2002) The International Journal, Gasification of hazelnut shells in a downdraft gasifier Vol. 27 pp 415-427.

19

scheme elements in this software. As for 2-D model the option is quadrilateral and

for 3-D hexahedron can be used. Boundary zones were created for examples inlet,

outlet, nozzle, throat, hot wall and axis This model finally exported to FLUENT 5.5

after the setting were specified to 2-D, axisymetric geometry and quadrilateral

meshing although these give few cells, high quality results can be obtained.

3.2.1.2 Solver

Solver is very important in solving the flow problem. FLUENT 5.5 have

many solvers and segregated (implicit) has been used to solve the direct combustion

process in gasifier. This is because it can solve both compressible and incompressible

flow problems. Besides that it can be used to solve the unsteady flows.

3.2.1.3 Post-processing

Post-processing is very significant in visualisation tool. Most flow properties

such as pressure, temperature, species concentration and density for the combustion

reaction can be displayed and analysed after simulation. Boundary conditions,

meshing and physical parameters are criterions that have to be revised if the solution

is not converging. The convergent solution is not essentially compulsory, but it is

inevitable since the most accurate solution is needed for post-processing. Grid

adaptive can be done in order to give more accuracy to the solution.

20

3.3 Model Setting

3.3.1 Dimension

The model’s setting in GAMBIT and FLUENT are discussed.

Figure 3.1: Schematic diagram of downdraft gasifier

Based on this typical diagram from pilot biomass gasifier that has been carried

out by Dogru et al. (2002), the model was drawn in GAMBIT software and then

imported to FLUENT code for further analysis. The model dimensions as shown

below:

• Height, H = 810 mm

• Diameter, D = 450 mm

• Throat diameter, d = 135 mm

• Fuel hopper, L = 305 mm

21

The boundary condition for each segments in the gasifier are specified in

FLUENT. FLUENT has a wide range of boundary conditions that permit flow to enter

and exit the solution domain. FLUENT provides 10 types of boundary cell types for

the specification of flow inlets and exits: velocity inlet, pressure inlet, mass flow inlet,

pressure outlet, pressure far-field, outflow, inlet vent, intake fan, outlet vent, and

exhaust fan. The inlet and exit boundary condition options in FLUENT that were

used in this 2-dimensional model are as follows:

· Velocity inlet boundary conditions are used to define the velocity and scalar

properties of the flow at inlet boundaries.

· Pressure inlet boundary conditions are used to define the total pressure and

other scalar quantities at flow inlets.

· Mass flow inlet boundary conditions are used in compressible flows to

prescribe a mass flow rate at an inlet. It is not necessary to use mass flow inlets in

incompressible flows because when density is constant, velocity inlet boundary

conditions will fix the mass flow.

· Pressure outlet boundary conditions are used to define the static pressure at

flow outlets (and also other scalar variables, in case of backflow). The use of a

pressure outlet boundary condition instead of an outflow condition often results in a

better rate of convergence when backflow occurs during iteration.

22

3.3.1.1 Downdraft gasifier

For atmospheric downdraft gasifier, the first settings for temperature are as

below

Expected temperature °C Model set-up°C

• Distillation zone 500 Fuel hopper 373

• Carbonisation zone 800 Hot-wall 800

• Oxidation zone 1200 Nozzle 737

• Reduction zone 300 Throat (outlet) 373

Model Settings Space 2D Time Steady Viscous Laminar Heat Transfer Enabled Melting-Freezing Disabled Radiation None

23

Species Transport Reacting (6 species) Coupled Dispersed Phase Disabled Pollutants Disabled Soot Disabled

After that each zone are classified as below for boundary condition: Zones

Name id Type Fluid 1 fluid Centre 2 symmetry Outlet 3 pressure-outlet Throat 4 wall Nozzle 5 velocity-inlet Outer-wall 6 wall Fuel-hopper 7 wall Default-interior 9 interior Relaxation

Variable Relaxation Factor Pressure 0.1 Momentum 0.1 Energy 0.89999998 Wood volatile 1 O2 1 CO 1 H2O 1 N2 1 Density 1 Body Forces 1

Discretization Scheme

Variable Scheme Pressure Standard Momentum First Order Upwind Pressure-Velocity Coupling SIMPLE Energy First Order Upwind

24

wood_vol First Order Upwind O2 First Order Upwind CO2 First Order Upwind H2O First Order Upwind

N2 First Order Upwind Material: wood (solid)

Property Units Method Value(s) Density kg/m3 constant 700 Cp (Specific Heat) j/kg-k constant 2310 Thermal Conductivity w/m-k constant 0.17299999

25

3.3.1.2 Pressurised Downdraft gasifier

Expected

Temperature, °C Model set-up,°C Pressure, MPa

• Distillation zone 500 Fuel hopper (inlet) 373 5.0

• Carbonisation zone 800 Hot-wall 800 -

• Oxidation zone 1200 Nozzle 737 -

• Reduction zone 300 Throat (outlet) 373 - The pressure is set to be 5.0 MPa at the fuel hopper. Here, the boundary condition is

changed to pressure inlet instead of wall in atmospheric downdraft gasifier.

26

3.4 Modelling Procedure

The diagram below illustrates the simplified procedures for the overall process in the

modelling and simulation.

Grid

Models

Materials properties

Boundary conditions

Solution

Displaying preliminary solution

Discretization

Adapting the grid

Iterations

Results

(Examples: xy Plots & Contour Plots for temperature & pressure) 27

Chapter 4 – Result Analysis & Discussion 4.1 Introduction In this chapter, the results are discussed and analysed between those two types of

gasifier; pressurised and atmospheric gasifier.

4.2 Model Description

The downdraft gasifier operates by heating the woodchips to around 830°C

(1526°F) using burner/torches that are connected at both side of cylinder. The

woodchips are heated in the reactor using air as the gasifying medium to produce a

syngas which contains varying amount of H2, CO, CH4, CO2 and N2. The outer wall

also are being applied a heat until the wood breaks, decomposes and gives off gas

which can be used as a fuel which is can be used in gas turbine. As for ignition

purposes and electricity generation, the syngas is cleaned and cooled prior to injection

into a spark ignition gas engine which is also fed with diesel support fuel.

28

Inlet

Outlet

Figure 4.1: Grids

Figure 4.1 shows the grids of biomass gasifier that was modelled in GAMBIT.

Only half of the gasifier was modelled due to the symmetrical geometry. Woodchips

are fed from the inlet and heat is applied through the hot wall and also from the

nozzle. The pressure and temperature of the gasifier are determined using the

simulation and compared with the result from experimental. The results are discussed

in the following topics.

29

Nozzle

Hot wall

Symmetrical axis

Throat

4.3 Result Analysis from downdraft gasifier

4.3.1 First-order Discretization.

In order to get the solution convergent, the setting of the model in FLUENT

has to be started from the First-order Discretization. Under-Relaxation Factors,

Energy is set to be 0.9. Simulation was run until around 4000 iterations to get the

convergent solution. The graph shows that the solution was not so smooth. There

were sinusoidal shapes in some lines of the solution. Therefore the setting was

changed to Second-order Discretization. The graph can be compared between both of

the settings.

Graph 4.1: Under-Relaxation Factors, Energy is set to be 0.9.

30

4.3.2 Second-order Discretization Second-order Discretization is used in order to improve the accuracy of the

solution. With Second-order Discretization, less aggressive (lower) values for energy

under-relaxation to ensure convergence was setting up. Thus, the energy was reduced

and was set up to 0.6. After 5500 iterations, the solution was convergent as show in

Graph 4.2; the line is smoother compare to the First-order Discretization.

Graph 4.2: Under-Relaxation Factors, Energy is set to be 0.6.

The simulation was continued with the changed in momentum setting. The setting for

momentum can be set to be small index number as well. The momentum is set to 0.6

and the solution was not converged. But for previous setting, the solution was

convergent as it was set to be 0.9. From the graph, it shows that the line is

continuously running until 25000 iterations.

31

Graph 4.3: Under-Relaxation Factors, Energy and Momentum are set to 0.6.

The gasifier solution can be improved further by refining the grid to be better

resolving the flow details. In this step, the adaptation of the grid based on the

temperature gradient in the current solution was used. The range of temperature

gradients over which to adapt was determined before refining the grid. The small

value was taken which was in the range between 0.02 to 0.01°C.

The contours will display the cell values of temperature instead of the smooth

looking node values. Here, the regions to adapt are significant to display in cell by

cell.

Comparison of the filled temperature contours for the first solution (using the

original grid and First-order Discretization) and the last solution (using an adapted

grid and Second-order Discretization) clearly indicate that the latter is much less

diffusive.

While the First-order Discretization is the default schema in FLUENT, it is

good to use first-order solution as a starting guess for a calculation that uses a higher-

order discretization scheme and, optionally, an adapted grid. In order to get more

efficient solution, to flow-field has to be compute first without solving the energy

solution and then solve the energy without solving the flow-field equations. 32

4.4 Downdraft gasifier contour results

4.4.1 Profile for first-order discretization. The temperature and pressure history of a particle in the downdraft gasifier

was the primary focus of a detailed characterisation study as this relationship in

theory could exclusively determine the extent to which the solid material reacts.

Particle temperature is influenced by radiative and convective transport, internal

conduction, chemical reaction and also most of it by hot wall impact.

Particle residence times are a function of the local gas flow rate and on the

path the particle takes inside the reactor which depends on particle density, gravity,

size, heat capacity and to some degree on particle starting location.

A simple model is not sufficient to capture the complexity of these problems.

CFD simulations were employed to model many of these complex processes as the

geometry and input condition can be accurately measured. This is because the model

boundary conditions are based measured quantities.

From the contour, the high molecular weight biomass pyrolysis vapours were

found to undergo thermal reactions at fairly moderate temperatures.

33

4.4.1.1 Temperature and pressure contour

1. Filled contour of temperature

Figure 4.2: Predicted total temperature after the Initial Calculation

According to the result form the simulation, the total temperature are mostly high

around the throat and nozzle. The temperature is higher at these areas because it is

affected by the burner/torches that is place at the wall near the throat. Furthermore,

these areas are oxidation zone and reduction zone where all the woodchips are burnt.

2. Filled contour of static pressure

From the contour of static pressure, there is not much different from predicted total

pressure. The pressure distributions are roughly the same. (Refer Figure 4.3 and

Figure 4.4)

34

Figure 4.3: Predicted Static Pressure Distribution after the Initial Calculation

3. Filled contour of total pressure

Figure 4.4: Predicted Total Pressure Distribution after the Initial Calculation

35

4.4.1.2 Temperature and pressure distribution

Graph 4.4: Total Temperature Distribution at the Throat

The graph above showed the total temperature inside the gasifer to the distance of the

gasifier along y-direction. 0 meter is the distance at the bottom of the gasifer which is

at the outlet. The distribution in Graph 4.4 is along the gasifier’s throat. The

temperature is increasing directly proportional to the distance.

36

Figure 4.5: Total Temperature Distribution at the Outlet

The temperature distribution at the outlet is decreasing with the distance. Some part of

the gasifier at the outlet has the constant temperature.

Figure 4.6: Total Pressure Distribution at the Outlet

The total pressure distribution at the outlet is decreasing gradually with the distance. 37

Graph 4.5: Total pressure at fuel hopper, nozzle, outer-wall, outlet and throat. Graph showed the total pressure at all the zones inside the gasifier. Total pressure is

constant at the fuel hopper. As for the outlet, the pressure is decreasing from the

bottom to the top of the gasifer. On the other hand the total pressure is increasing for

the area around the nozzle.

38

4.5 Result Analysis from pressurised downdraft gasifier

4.5.1 First-order Discretization.

As for pressurized downdraft gasifier, simulation was run only until 300 iterations in

order to get the convergent solution. From the residual graph above, the solution was

not stable as some simulation stopped at small iteration which is less than 100

iterations. Therefore the setting in FLUENT software was changed to Second-order

Discretization. Unfortunately, the graph was almost the same. Both are not stable.

4.5.2 Second-order Discretization

After the settings were changed to second-order discretization, the residual graph was

almost the same which is not stable.

39

4.6 Pressurised downdraft gasifier contour results

4.6.1 Profile for first-order Discretization

4.6.1.1 Temperature and pressure contour

1. Filled contour of static pressure

Figure 4.7: Predicted Static Pressure Distribution after the Initial Calculation From the filled contour of static and total pressure (Refer Graph 4.7 & 4.8), the graph

show that at the top of the gasifier, the pressure is high which is at the fuel hopper.

Both graph are almost the same.

40

2. Filled contour of total pressure

Figure 4.8: Predicted Total Pressure Distribution after the Initial Calculation 3. Filled contour of total temperature

Figure 4.9: Predicted Total Temperature Distribution after the Initial Calculation

From the contour, the total temperature obtained shows no difference of temperature

inside the gasifier.

41 4.6.1.2 Temperature and pressure distribution

Graph 4.6: Total pressure distribution at outlet

The total pressure distributions at outlet show almost the same for atmospheric and

pressurised gasifier.

42

Graph 4.7: Total pressure at throat

The graph above shows the total pressure distribution inside the gasifier to the

distance along y-direction. It can be seen that the pressure is increasing from 1 x 105

Pascal to 1 x 106 Pascal.

Graph 4.8: Total pressure at fuel hopper, nozzle, outer-wall, outlet and throat. 43

The graph above shows the pressure at all the segment in the biomass gasifier. From

the fuel-hopper to the throat, the pressure drop is recorded from 2.5 x 106 Pascal to 2.5

x 105 Pascal.

44

Chapter 5 – Conclusion & Recommendations

The computational fluid dynamics (CFD) software is playing a strong role as a

design tool computer. Today CFD can be considered as an equal partner with pure

theory and pure experiment in the analysis and solution of fluid dynamics problems in

dealing with fluid flow. The development of new products of gasifer for biomass is

getting better if the trial and error methods are replaced with CFD software which can

assist them in improving new products.

In the dissertation, some of the result obtained from this simulation is not very

accurate. This is because of several factors such as model’s parameters, geometry,

meshing, grid and discretization that play an important role in the result accuracy.

As for model’s geometry, 2-dimensional model is used because of easy

iterations and time consuming. Thus, first-order discretization leads to convergent

result. Second-order discretization is useful to improve the result accuracy. After

some iteration, the result was not converged. Therefore second-order discretization is

not suitable for 2-dimensional model20.

In the first method used as a preliminary approach, the combustion process is

assumed to be laminar and steady state. In real life, combustion process is involved

with turbulent and unsteady state processes. Therefore, further study is needed on

these processes.

There is consideration that has to put in mind about the environmental impact,

from indirect and direct emissions release from the combustion process. For this

dissertation, universal emission-releasing gas model based on biomass fuel is not

evaluated although it is very necessary to analyse the product gas from the

combustion process. This is because each case-dependent emissions associated with

biomass gasification practice is quite tedious and time-consuming.

Thus, it is necessary in future to model the pollutants gas so that it can be used

as reference when any gasifier are to be built on local community or society. So that

they can be modelled based on simulation and can be evaluated in advance, as the

result can be demonstrated from modelling using computational fluid dynamic

software.

20 Anderson, J.D (1995), Computational Fluid Dynamics, The Basic With Application, International Editions, McGraw-Hill

45

Finally, the software FLUENT is used and most of the work is directed

towards combustion and gasification of biomass fuels. The number of persons

involved in simulations is such that the computational capacity is still a limiting

factor. The available workstations will as far as possible be used for pre-processing

and post-processing of the problems and for tests on more coarse grids. But to be able

to fulfil the real project in specified time, it is essential to get access to high

computational capacity and extended memory.

46

References

1. Ronald, V.S. (2002), How European Waste Will Contribute to Renewable Energy.

Vol. 30 pp. 471-475.

2. Dumbleton, F. (2001), Standardisation of Solid Biofuels in the UK , AEA

Technology Environment, Report 1, ESTA 32192001.

3. McIlveen, W. et al. (2001), Bioresource Technology, A re-appraisal of wood-fired

combustion, Vol. 76, pp 183-190

4. Dogru, M. and et al (2002), The International Journal, Gasification of hazelnut

shells in a downdraft gasifier Vol. 27 pp 415-427.

5. Dogru, M et al. (2002), Fuel Processing Technology, Gasification of sewage sludge

using throated downdraft gasifier and uncertainty analysis, Vol. 75, pp. 55-82.

6. Richard, L.B. (1996), Biomass-Fired Power Generation in National Renewable

Energy Laboratory Conference, New York: Engineering Foundation.

7. Purnomo, D.J.A & Ragland, K.W. (1990), Twenty-Third Symposium

(International) on Combustion, Pressurised combustion of woodchips, pp 1025-1032.

8. Xiu, L.Y. et al (2002), Biomass & Bioenergy, Design and operation of CFB

gasification and power generation system for rice husk, Vol. 23, pp 181-187.

9. Heermann, C et al. (2000), Pyrolysis & Gasification of Waste. 2nd ed. Juniper

Consultancy Services Ltd.

10. Miller, R.S & Bellan, J. (1997), Combustion Science Technology, A generalised

biomass pyrolysis model based on superimposed cellulose, hemicellulose and lignin

kinetis, No. 126, pp 97-137.

47

11. Purnomo, D.J.A & Ragland, K.W. (1990), Twenty-Third Symposium

(International) on Combustion, Pressurised combustion of woodchips, pp 1025-1032.

12. Eastop, T.D and McConkey, A. (1998), Applied Thermodynamics for Engineering

Technologist, 5th ed, Addison Wesly Longman Limited.

13. LIOR International LV. (2002), Biomass Combustion Technology (7) Wood Lower Calorific Value http://www.lior-int.com/FORMULA/rp7.as

14. Robinson, A et al. (1998), Fireside Issues Associated with Coal-Biomass Co-

firing, in Conference on BioEnergy ’98, United State: National Renewable Energy

Laboratory.

15. McIlveen, W. et al. (2001), Bioresource Technology, A re-appraisal of wood-fired

combustion, Vol. 76, pp 183-190

16. Brink, A et al. (2001), IFRF Combustion Journal, Modelling Nitrogen Chemistry

in the Freeboard of Biomass-FBC, pp 1-14.

17. Whalen, P. (2000), Fluent Inc. Releases Version 5.5 of Its FLUENT Software,

http://www.fluent.com/about/news/pr/pr17.htm

18. Dogru, M. and et al (2002), The International Journal, Gasification of hazelnut

shells in a downdraft gasifier Vol. 27 pp 415-427.

19. Anderson, J. D (1995), Computational Fluid Dynamics, The Basic With

Application, International Editions, McGraw-Hill.

48

Appendices Appendix 1.1: List of gasifier Downdraft gasifier, biomass fuel is fed in at the top, gas is

sucked off near the base, and air is drawn down through the

fuel.

• Size, shape and moisture content of the biomass

particles must be controlled within close limits.

• Quality of the gas fuel produced is generally good.

• The downdraft principle is only suitable for

installations of less than about 1 MW electrical

output.

Updraft gasifier is one where the gas flows upwards as the

biomass moves down.

• The updraft principle is suitable for installations of a

few tens of megawatt capacity.

• Size, shape and moisture content of the biomass

particles are much less critical than with a downdraft

gasifier.

• Quality of the gas produced tends to be poor.

Other types of fixed bed gasifier, produced in small numbers, may involve:

• cross flow design; • use of oxygen as the gasifying agent instead of air; • Feeding from the bottom by means of a transport screw.

Fluidised bed gasifiers have either a bubbling bed or a circulating bed.

• There is no technical scale-up limit (though size may be limited by availability

of biomass or the local energy demand).

49

• The gasifying agent is usually air at atmospheric pressure, but pressurised

gasification can be advantageous when supplying gas turbines larger than

about 100 MW.

• The danger of ash agglomeration is low, because of the relatively low

temperature (about 850°C).

50

Appendix 1.2: Summary – First-order Discretization for downdraft

gasifier FLUENT Version: 2d, segregated, spe6, lam (2d, segregated, 6 species, laminar) Release: 5.5.14 Title: Gasifier Models ------ Model Settings ---------------------------------------------- Space 2D Time Steady Viscous Laminar Heat Transfer Enabled Melting-Freezing Disabled Radiation None Species Transport Reacting (6 species) Coupled Dispersed Phase Disabled Pollutants Disabled Soot Disabled Boundary Conditions ------------------- Zones Name id type --------------------------------------- Fluid 1 fluid Centre 2 symmetry Outlet 3 pressure-outlet Throat 4 wall Nozzle 5 velocity-inlet Outer-wall 6 wall Fuel-hopper 7 wall Default-interior 9 interior Boundary Conditions Fluid Condition Value --------------------------------------------------------------------- Material Name wood-volatiles-air Specify source terms? no Source Terms ((mass (constant . 0) (profile )) (x-momentum (constant . 0) (profile )) (y-momentum (constant . 0) (profile )) (energy (constant . 0) (profile )) (species-0 (constant . 0) (profile )) (species-1 (constant . 0) (profile )) (species-2 (constant . 0) (profile )) (species-3 (constant . 0) (profile )) (species-4 (constant . 0) (profile )))

51

Motion Type 0 X-Velocity Of Zone 0

Y-Velocity Of Zone 0 Rotation speed 0 X-Origin of Rotation-Axis 0 Y-Origin of Rotation-Axis 0 Porous zone? no X-Component of Direction-1 Vector 1 Y-Component of Direction-1 Vector 0 Direction-1 Viscous Resistance 0 Direction-2 Viscous Resistance 0 Direction-3 Viscous Resistance 0 Direction-1 Inertial Resistance 0 Direction-2 Inertial Resistance 0 Direction-3 Inertial Resistance 0 C0 Coefficient for Power-Law 0 C1 Coefficient for Power-Law 0 Porosity 1

Solid Material Name aluminum Centre Condition Value ----------------- Outlet Condition Value --------------------------------------------------------------------- Gauge Pressure 0 Backflow Total Temperature 300

Backflow (((constant . 0.77993202) (profile )) ((constant . 0.22006799) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile ))) Throat Condition Value --------------------------------------------------------------------- Wall Thickness 0 Heat Generation Rate 0 Material Name aluminium Thermal BC Type 0 Temperature 1473 Heat Flux 0 Convective Heat Transfer Coefficient 0 Free Stream Temperature 300 Apply a velocity to this wall? no Define wall motion relative to adjacent cell zone? yes Apply a rotational velocity to this wall? no Velocity Magnitude 0 X-Component of Wall Translation 1 Y-Component of Wall Translation 0 Define wall velocity components? no X-Component of Wall Translation 0 Y-Component of Wall Translation 0

52

External Emissivity 1 External Radiation Temperature 300 (1 1 0 0 0) (((constant . 0.77993202) (profile )) ((constant . 0.22006799) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )))

Rotation Speed 0 X-Position of Rotation-Axis Origin 0 Y-Position of Rotation-Axis Origin 0 Specify shear stress? no X-component of shear stress 0 Y-component of shear stress 0 Nozzle Condition Value --------------------------------------------------------------------- Velocity Specification Method 2 Reference Frame 0 Velocity Magnitude 2 X-Velocity 0 Y-Velocity 0 X-Component of Flow Direction 1 Y-Component of Flow Direction 0 X-Component of Axis Direction 1 Y-Component of Axis Direction 0 Z-Component of Axis Direction 0 X-Coordinate of Axis Origin 0 Y-Coordinate of Axis Origin 0 Z-Coordinate of Axis Origin 0 Angular velocity 0 Temperature 1273 (((constant . 0.77993202) (profile )) ((constant . 0.22006799) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile ))) Outer-wall Condition Value --------------------------------------------------------------------- Wall Thickness 0 Heat Generation Rate 0 Material Name aluminum Thermal BC Type 0 Temperature 620 Heat Flux 0 Convective Heat Transfer Coefficient 0 Free Stream Temperature 300 Apply a velocity to this wall? no Define wall motion relative to adjacent cell zone? yes Apply a rotational velocity to this wall? no Velocity Magnitude 0 X-Component of Wall Translation 1 Y-Component of Wall Translation 0 Define wall velocity components? no X-Component of Wall Translation 0

53

Y-Component of Wall Translation 0 External Emissivity 1 External Radiation Temperature 300 (1 1 0 0 0) (((constant . 0.77993202) (profile )) ((constant . 0.22006799) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile ))) Rotation Speed 0 X-Position of Rotation-Axis Origin 0 Y-Position of Rotation-Axis Origin 0 Specify shear stress? no X-component of shear stress 0 Y-component of shear stress 0 Fuel-hopper Condition Value --------------------------------------------------------------------- Wall Thickness 0 Heat Generation Rate 0 Material Name aluminium Thermal BC Type 0 Temperature 343 Heat Flux 0 Convective Heat Transfer Coefficient 0 Free Stream Temperature 300 Apply a velocity to this wall? no Define wall motion relative to adjacent cell zone? yes Apply a rotational velocity to this wall? no Velocity Magnitude 0 X-Component of Wall Translation 1 Y-Component of Wall Translation 0 Define wall velocity components? no X-Component of Wall Translation 0 Y-Component of Wall Translation 0 External Emissivity 1 External Radiation Temperature 300 (1 1 0 0 0) (((constant . 0.77993202) (profile )) ((constant . 0.22006799) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile ))) Rotation Speed 0 X-Position of Rotation-Axis Origin 0 Y-Position of Rotation-Axis Origin 0 Specify shear stress? no X-component of shear stress 0 Y-component of shear stress 0 Default-interior Condition Value -----------------

54

Solver Controls --------------- Equations Equation Solved ----------------- Flow yes Energy yes wood_vol yes o2 yes co2 yes h2o yes n2 yes Numerics Numeric Enabled --------------------------------------- Absolute Velocity Formulation yes Relaxation Variable Relaxation Factor ------------------------------- Pressure 0.1 Momentum 0.1 Energy 0.89999998 wood_vol 1 o2 1 co2 1 h2o 1 n2 1 Density 1 Body Forces 1 Linear Solver Solver Termination Residual Reduction Variable Type Criterion Tolerance -------------------------------------------------------- Pressure V-Cycle 0.1 X-Momentum Flexible 0.1 0.7 Y-Momentum Flexible 0.1 0.7 Energy Flexible 0.1 0.7 wood_vol Flexible 0.1 0.7 o2 Flexible 0.1 0.7 co2 Flexible 0.1 0.7 h2o Flexible 0.1 0.7 n2 Flexible 0.1 0.7 Discretization Scheme Variable Scheme ----------------------------------------------- Pressure Standard Momentum First Order Upwind Pressure-Velocity Coupling SIMPLE Energy First Order Upwind wood_vol First Order Upwind

55

o2 First Order Upwind co2 First Order Upwind h2o First Order Upwind n2 First Order Upwind Solution Limits Quantity Limit ----------------------------------- Minimum Absolute Pressure 1 Maximum Absolute Pressure 5000000 Minimum Temperature 1 Maximum Temperature 5000 Material Properties ------------------- Material: aluminium (solid) Property Units Method Value(s) --------------------------------------------------- Density kg/m3 constant 2719 Cp (Specific Heat) j/kg-k constant 871 Thermal Conductivity w/m-k constant 202.4 Material: air (fluid) Property Units Method Value(s) ----------------------------------------------------------- Molecular Weight kg/kgmol constant 28.966 Standard State Enthalpy j/kgmol constant 0 Standard State Entropy j/kgmol-k constant 0 Reference Temperature k constant 298.15 L-J Characteristic Length angstrom constant 3.711 L-J Energy Parameter k constant 78.6 Degrees of Freedom constant 0 Material: wood (solid) Property Units Method Value(s) ----------------------------------------------------- Density kg/m3 constant 700 Cp (Specific Heat) j/kg-k constant 2310 Thermal Conductivity w/m-k constant 0.17299999 Material: wood-volatiles (fluid) Property Units Method Value(s) ------------------------------------------------------------- Molecular Weight kg/kgmol constant 31.626 Standard State Enthalpy j/kgmol constant -2.548e+08 Standard State Entropy j/kgmol-k constant 0 Reference Temperature k constant 298.15 L-J Characteristic Length angstrom constant 0 L-J Energy Parameter k constant 0 Degrees of Freedom constant 0

56

Material: oxygen (fluid)

Property Units Method Value(s) ------------------------------------------------------------ Molecular Weight kg/kgmol constant 31.9988 Standard State Enthalpy j/kgmol constant 0 Standard State Entropy j/kgmol-k constant 205026.86 Reference Temperature k constant 298.15 L-J Characteristic Length angstrom constant 3.458 L-J Energy Parameter k constant 107.4 Degrees of Freedom constant 0 Material: carbon-dioxide (fluid) Property Units Method Value(s) --------------------------------------------------------------- Molecular Weight kg/kgmol constant 44.00995 Standard State Enthalpy j/kgmol constant - 3.9353235e+08 Standard State Entropy j/kgmol-k constant 213715.88 Reference Temperature k constant 298.15 L-J Characteristic Length angstrom constant 3.941 L-J Energy Parameter k constant 195.2 Degrees of Freedom constant 0 Material: water-vapor (fluid) Property Units Method Value(s) --------------------------------------------------------------- Molecular Weight kg/kgmol constant 18.01534 Standard State Enthalpy j/kgmol constant -2.418379e+08 Standard State Entropy j/kgmol-k constant 188696.44 Reference Temperature k constant 298.15 L-J Characteristic Length angstrom constant 2.605 L-J Energy Parameter k constant 572.4 Degrees of Freedom constant 0 Material: nitrogen (fluid) Property Units Method Value(s) ------------------------------------------------------------ Molecular Weight kg/kgmol constant 28.0134 Standard State Enthalpy j/kgmol constant 0 Standard State Entropy j/kgmol-k constant 191494.78 Reference Temperature k constant 298.15 L-J Characteristic Length angstrom constant 3.621 L-J Energy Parameter k constant 97.53 Degrees of Freedom constant 0

57

Material: wood-volatiles-air (mixture) Property Units Method Value(s) --------------------------------------------------------------------- Mixture Species names ((wood_vol o2 co2 h2o n2 air) ()) Reaction Model finite-rate ((reaction-1 ((wood_vol 1 0.2 1) (o2 1.058 1.3 1) (n2 1 1 1)) ((co2 1 0 1) (h2o 1.191 0 1) (n2 1 0 1)) ((air 0 1)) (stoichiometry 1wood_vol + 1.058o2 + 1n2 --> 1co2 + 1.191h2o + 1n2) (arrhenius 2.1190001e+11 2.027e+08 0) (mixing-rate 4 0.5) (use-third-body-efficiencies? . #f) (surface-reaction? . #f) (backward-reaction? . #f)))

Density kg/m3 incompressible-ideal-gas #f Cp (Specific Heat) j/kg-k constant 1000 Thermal Conductivity w/m-k constant 0.045400001 Viscosity kg/m-s constant 1.72e-05 Mass Diffusivity m2/s constant-dilute-appx (2.88e-05) Thermal Diffusion Coefficient kg/m-s kinetic-theory #f Thermal Expansion Coefficient 1/k constant 0

58

Appendix 1.3: Summary – First-order Discretization for pressurised

downdraft gasifier FLUENT Version: 2d, segregated, spe6, lam (2d, segregated, 6 species, laminar) Release: 5.5.14 Title: Models ------ Model Settings ---------------------------------------------- Space 2D Time Steady Viscous Laminar Heat Transfer Enabled Melting-Freezing Disabled Radiation None Species Transport Reacting (6 species) Coupled Dispersed Phase Disabled Pollutants Disabled Soot Disabled Boundary Conditions ------------------- Zones Name id type --------------------------------------- Fluid 1 fluid Fuel-hopper 7 pressure-inlet Centre 2 symmetry Outlet 3 pressure-outlet Throat 4 wall Nozzle 5 velocity-inlet Outer-wall 6 wall Default-interior 9 interior Boundary Conditions Fluid Condition Value --------------------------------------------------------------------- Material Name wood-volatiles-air Specify source terms? no Source Terms ((mass (constant . 0) (profile )) (x-momentum (constant . 0) (profile )) (y-momentum (constant . 0) (profile )) (energy (constant . 0) (profile )) (species-0 (constant . 0) (profile )) (species-1 (constant . 0) (profile )) (species-2 (constant . 0) (profile )) (species-3 (constant . 0) (profile )) (species-4 (constant . 0) (profile )))

59

Motion Type 0 X-Velocity Of Zone 0 Y-Velocity Of Zone 0 Rotation speed 0 X-Origin of Rotation-Axis 0 Y-Origin of Rotation-Axis 0 Porous zone? No X-Component of Direction-1 Vector 1 Y-Component of Direction-1 Vector 0 Direction-1 Viscous Resistance 0 Direction-2 Viscous Resistance 0 Direction-3 Viscous Resistance 0 Direction-1 Inertial Resistance 0 Direction-2 Inertial Resistance 0 Direction-3 Inertial Resistance 0 C0 Coefficient for Power-Law 0 C1 Coefficient for Power-Law 0 Porosity 1 Solid Material Name aluminium Fuel-hopper Condition Value --------------------------------------------------------------------- Gauge Total Pressure 2500000 Supersonic/Initial Gauge Pressure 0 Total Temperature 373 Direction Specification Method 1 X-Component of Flow Direction 1 Y-Component of Flow Direction 0 X-Component of Axis Direction 0 Y-Component of Axis Direction 0 Z-Component of Axis Direction 1 X-Coordinate of Axis Origin 0 Y-Coordinate of Axis Origin 0 Z-Coordinate of Axis Origin 0 (((constant . 0.77993202) (profile )) ((constant . 0.22006799) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile ))) Centre Condition Value ----------------- Outlet Condition Value --------------------------------------------------------------------- Gauge Pressure 0 Backflow Total Temperature 300 Backflow (((constant . 0.77993202) (profile )) ((constant . 0.22006799) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )))

60

Throat Condition Value --------------------------------------------------------------------- Wall Thickness 0 Heat Generation Rate 0 Material Name aluminium Thermal BC Type 0 Temperature 373 Heat Flux 0 Convective Heat Transfer Coefficient 0 Free Stream Temperature 300 Apply a velocity to this wall? no Define wall motion relative to adjacent cell zone? yes Apply a rotational velocity to this wall? no Velocity Magnitude 0 X-Component of Wall Translation 1 Y-Component of Wall Translation 0 Define wall velocity components? no X-Component of Wall Translation 0 Y-Component of Wall Translation 0 External Emissivity 1 External Radiation Temperature 300 (1 1 0 0 0) (((constant . 0.77993202) (profile )) ((constant . 0.22006799) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile ))) Rotation Speed 0 X-Position of Rotation-Axis Origin 0 Y-Position of Rotation-Axis Origin 0 Specify shear stress? no X-component of shear stress 0 Y-component of shear stress 0 Nozzle Condition Value --------------------------------------------------------------------- Velocity Specification Method 2 Reference Frame 0 Velocity Magnitude 2 X-Velocity 0 Y-Velocity 0 X-Component of Flow Direction 1 Y-Component of Flow Direction 0 X-Component of Axis Direction 1 Y-Component of Axis Direction 0 Z-Component of Axis Direction 0 X-Coordinate of Axis Origin 0 Y-Coordinate of Axis Origin 0 Z-Coordinate of Axis Origin 0 Angular velocity 0 Temperature 737 (((constant . 0.77993202) (profile )) ((constant . 0.22006799) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )))

61

Outer-wall Condition Value --------------------------------------------------------------------- Wall Thickness 0 Heat Generation Rate 0 Material Name aluminium Thermal BC Type 0 Temperature 800 Heat Flux 0 Convective Heat Transfer Coefficient 0 Free Stream Temperature 300 Apply a velocity to this wall? no Define wall motion relative to adjacent cell zone? yes Apply a rotational velocity to this wall? no Velocity Magnitude 0 X-Component of Wall Translation 1 Y-Component of Wall Translation 0 Define wall velocity components? no X-Component of Wall Translation 0 Y-Component of Wall Translation 0 External Emissivity 1 External Radiation Temperature 300 (1 1 0 0 0) (((constant . 0.77993202) (profile )) ((constant . 0.22006799) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile )) ((constant . 0) (profile ))) Rotation Speed 0 X-Position of Rotation-Axis Origin 0 Y-Position of Rotation-Axis Origin 0 Specify shear stress? no X-component of shear stress 0 Y-component of shear stress 0 Default-interior Condition Value ----------------- Solver Controls --------------- Equations Equation Solved ----------------- Flow yes Energy yes wood_vol yes o2 yes co2 yes h2o yes n2 yes

62